General project discussion.
Use this as a last resort if your topic does not fit in any of the other boards but it still on-topic.
Forum rules
This General Discussion board is meant for topics that are still relevant to Pale Moon, web browsers, browser tech, UXP applications, and related, but don't have a more fitting board available.
Please stick to the relevance of this forum here, which focuses on everything around the Pale Moon project and its user community. "Random" subjects don't belong here, and should be posted in the Off-Topic board.
-
Daikun
- Lunatic
- Posts: 443
- Joined: 2013-12-13, 20:54
- Location: California
Unread post
by Daikun » 2020-12-21, 08:10
Version 28 has to be the longest-lasting version of Pale Moon we've seen so far. We're up to 28.17 now?
I'm amazed we've gone this far in the double-digits. We'd have been lucky to see x.9.0 in previous versions of the browser.
How much farther do you plan to take PM28? Will we see version 29 anytime in the future?
-
adesh
- Board Warrior
- Posts: 1277
- Joined: 2017-06-06, 07:38
Unread post
by adesh » 2020-12-21, 08:36
Daikun wrote: ↑2020-12-21, 08:10
Will we see version 29 anytime in the future?
Hopefully yes. Version 29 is blocked by WebComponents shit.
-
Moonchild
- Pale Moon guru
- Posts: 35600
- Joined: 2011-08-28, 17:27
- Location: Motala, SE
-
Contact:
Unread post
by Moonchild » 2020-12-21, 09:05
Not just that, extensions have also been a problem as we couldn't just bump the version to 29 when Firefox extensions were still somewhat accepted, because of expected issues with extensions catering to Australis when seeing v29 internally. So we've decided to let it run into the double digits, instead.
"Sometimes, the best way to get what you want is to be a good person." -- Louis Rossmann
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite
-
New Tobin Paradigm
Unread post
by New Tobin Paradigm » 2020-12-21, 09:46
Take it as precious extra time for old and insecure unmodified Firefox extensions.
Still need WebComponents though.
-
moonbat
- Knows the dark side
- Posts: 4980
- Joined: 2015-12-09, 15:45
-
Contact:
Unread post
by moonbat » 2020-12-21, 11:51
Of late, I've seen a lot of snark (on the subreddit, where else) over the 'Your browser your way' slogan with reference to not agreeing to support legacy Firefox extensions as-is till the heat death of the universe.
Given that Basilisk self identifies as Firefox and is compatible with version 52, would it be accurate to direct these people to it instead? At which point they bring up how Waterfox Classic is so much better for supporting both XUL and WebExtensions
-
Lunokhod
- Lunatic
- Posts: 469
- Joined: 2017-04-20, 21:25
-
Contact:
Unread post
by Lunokhod » 2020-12-21, 16:44
The unstable version I'm using now is 29.0.0a6 (and working very well) so you can easily get a "29" plate model if you want.
Given semantic versioning surely the long time on 28 is a good thing, it means nothing has been broken for users in updates:
"Major version X (X.y.z | X > 0) MUST be incremented if any backwards incompatible changes are introduced to the public API. It MAY also include minor and patch level changes. Patch and minor version MUST be reset to 0 when major version is incremented."
Wait, it's all Ohio? Always has been...
-
Moonchild
- Pale Moon guru
- Posts: 35600
- Joined: 2011-08-28, 17:27
- Location: Motala, SE
-
Contact:
Unread post
by Moonchild » 2020-12-21, 17:13
Lunokhod wrote: ↑2020-12-21, 16:44
"Major version X (X.y.z | X > 0) MUST be incremented if any backwards incompatible changes are introduced to the public API. It MAY also include minor and patch level changes. Patch and minor version MUST be reset to 0 when major version is incremented."
We do not adhere to any formal statement like that and we HAVE introduced backwards incompatible changes to "the public API" (considering we don't have a formal API to begin with, what the heck are you quoting!?) insofar as extension compatibility has changed. Also, our numbering is milestone.major.minor/sec.OOB, so the second point version is our major version.
Please don't try to be an authority on something you know jack shit about.
"Sometimes, the best way to get what you want is to be a good person." -- Louis Rossmann
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite
-
MGB1971
- Moonbather
- Posts: 53
- Joined: 2018-05-11, 09:31
Unread post
by MGB1971 » 2020-12-21, 19:06
Does it really matter what 'version' an update is?
An update is an update is further progress. I would be quite happy, if Moonchild so decided, to see Pale Moon 28.90.
-
Lunokhod
- Lunatic
- Posts: 469
- Joined: 2017-04-20, 21:25
-
Contact:
Unread post
by Lunokhod » 2020-12-21, 19:25
Well I guess not everyone knows about the widely used and accepted international standards for assigning version numbers then:
https://semver.org/
https://en.wikipedia.org/wiki/Software_versioning
Although not everyone uses the scheme yet and it's not compulsory of course. Most Linux package versions use the 3 part number in that manner, as it helps with system maintenance, so I just tend to look at a 3 part version number in those terms, apparently not the case here despite it's layout.
Wait, it's all Ohio? Always has been...
-
Moonchild
- Pale Moon guru
- Posts: 35600
- Joined: 2011-08-28, 17:27
- Location: Motala, SE
-
Contact:
Unread post
by Moonchild » 2020-12-21, 19:44
Look, I don't care what a website says about version numbering, and i certainly don't appreciate you coming off as if you have any authority over it with quoting RFC-style with MUST and other compelling terminology.
There is no standard for version numbers, period. They are by their very nature arbitrary and
entirely the developer's choice. We have our versioning for a number of very good reasons.
It's fine if you/someone want/wants to propose guidelines but taking it into MUST territory is taking it a few steps too far. Once you create your own software product you can be as strict as you want, but don't go applying it willy-nilly to everything.
Also, I've clearly written something up about this already:
viewtopic.php?f=24&t=6505
EDIT: mid-aired with coffebreak.
"Sometimes, the best way to get what you want is to be a good person." -- Louis Rossmann
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite
-
New Tobin Paradigm
Unread post
by New Tobin Paradigm » 2020-12-21, 20:02
Why don't you go complain to Mozilla about them getting mighty close to version 100 .. else come to my board and I will argue why I use a build number and typically have stopped increasing the First and Second version point.
Why don't you complain about ubuntu's date based version number or any of the other schemes out there like odd being dev and even being stable. At teh risk of this being thrown back at me.. Not everything has only one possible way.. There are many ways some better than others. So far you haven't even come to us with a compelling argument why one system is better than another only we should follow this because it seems most prevalent or because you personally like it.
Of course we could just stop showing the version number all together if you want.
As an aside, we merely only need to be consistent with the behavior of nsIVersionComparitor which is a shit load more flexible about evaluating version numbers than most schemes. Hell I even have a php port of it.
-
Lunokhod
- Lunatic
- Posts: 469
- Joined: 2017-04-20, 21:25
-
Contact:
Unread post
by Lunokhod » 2020-12-21, 20:20
It wasn't me that wrote MUST, that's just a quote from the standard. And there is a software API, thinking about it - the interface for extensions, and UXP itself. According to the standard there's also some latitude as it suggests you shouldn't change the major version too often so you end up at a large number quickly. So the fact Pale Moon uses UXP and supports addons while Firefox has dropped that and now bases itself on Chrome while changing it's major release number frequently is showing that Pale Moon is doing something good by not following suit, the lack of major version changes should only indicate it still works as well as ever and has not broken anything (much) which would be fairly equivalent to semantic versioning anyway.
But I see you have already laid out a version standard where the most significant number is the Milestone, I apologise for my misunderstanding. Not so much a question of how much farther can it go then - but why has it taken 2 years to move less than a mile?
Probably your version scheme was implemented before the semantic versioning became common, and some browsers do tend to use bigger version numbers to try and sound more impressive as a marketing ploy.
Wait, it's all Ohio? Always has been...
-
New Tobin Paradigm
Unread post
by New Tobin Paradigm » 2020-12-21, 20:46
The answer to your query is obviously: "Whatever, whatever, I do what I want".
-
Kendo
Unread post
by Kendo » 2020-12-22, 04:30
Moonchild wrote: ↑2020-12-21, 19:44
There is no standard for version numbers...
There is but it is being ignored. Chrome started incrementing the major version to suggest that their browser was more advanced/advancing faster than other browsers like Firefox. Naturally Firefox, not wanting to be seen as "old" followed the chase.
But generally speaking an update for version 1.2.3.4 that may include security fixes with no dramatic changes to compatibility, should only increment the minor number from 4 to 5, or if something has changed that may affect add-ons/plugins then middle numbers of 2 and 3 might be incremented. But the major number (1) should not be incremented unless the core has changed, making it incompatible with most add-ons/plugins, such as when PM migrated to XUL.
Most in the software industry follow the rule.
-
adesh
- Board Warrior
- Posts: 1277
- Joined: 2017-06-06, 07:38
Unread post
by adesh » 2020-12-22, 04:35
Off-topic:
This discussion is not fruitful. Please stop continuing it.
-
New Tobin Paradigm
Unread post
by New Tobin Paradigm » 2020-12-22, 04:50
I agree.
-
Daikun
- Lunatic
- Posts: 443
- Joined: 2013-12-13, 20:54
- Location: California
Unread post
by Daikun » 2021-02-03, 00:04
Daikun wrote: ↑2020-12-21, 08:10
Will we see version 29 anytime in the future?
Well, this came along quicker than I expected.